Diagnosing and staging bile duct cancer: cholangiographically guided biopsies or cholangioscopy?

In this issue of Endoscopy, two articles focus on the diagnostic utility of advanced techniques to diagnose cholangiocarcinoma and to assess the extent of malignancy to guide surgical resection.

Jahng et al. address the long-standing issue of improving the modest yield of intraluminal tissue biopsy of indeterminate bile duct strictures [1]. In a set of 61 patients they retrospectively tested 89 biopsies for histology and stained tissue specimens for p53, Ki-67, carcinoembryonic antigen (CEA), CA19–9, CAM5.2 and intracytoplasmic lumina (ICL). Histology alone revealed a sensitivity of 53 % and a specificity of 100 %. Addition of CAM5.2 or ICL increased sensitivity to 60 % and 73 %, respectively, with almost no compromise of specificity and positive predictive value (PPV). The two stains in combination did not significantly increase sensitivity. Other stains also increased sensitivity, but at the cost of lowering specificity and PPV. The negative predictive value (NPV) of markers alone or in combination was poor. In their discussion the authors list some potential limitations of their study such as the inclusion of repeat biopsies and absence of gold standard tissue diagnosis in all patients. To this can be added that the proportions of patients (or samples) are somewhat arbitrarily chosen with approximately 50 % being benign strictures; this influences PPV and NPV, and may not be representative of daily practice in other centers. Another point not explicitly stated in the manuscript is whether all tests were performed on adjacent tissue cuts from single biopsy samples. Most importantly, these results were obtained in a retrospective series, and it is well known that when novel tests are applied in clinical practice on the basis of promising results from retrospective series, the result is often disappointing. Therefore, before any definite conclusion can be drawn these results should be validated in a prospective series. Nevertheless, if these tests truly improve the sensitivity of intraductal biopsies and, at modest costs, detect more patients with malignancy, they should be considered of added value. On the other hand, from a practical and patient-management point of view, a test with a sensitivity of 75 % still misses 25 % of cases and, most importantly, a negative test result does not rule out disease with any certainty.

Kawakami et al. describe the use and yield of peroral cholangioscopy (POCS), with and without mapping biopsies, to diagnose bile duct carcinoma and to assess the extent of proximal and distal intraepithelial tumor spread (ITS) to guide surgical resection [2]. This retrospective study included 44 patients with localized-type bile duct carcinoma (LBDC). LBDC is classified as an elevated lesion with mild extramural invasion, but with significant ITS alongside the bile ducts both distal and proximal to the primary tumor (> 2 cm) in 20 % to 30 % of cases [3] [4]. This is in contrast to diffuse sclerosing bile duct cancer in which extramural invasion and submucosal tumor spread are more common. The utility of POCS with and without mapping biopsies was compared with the diagnostic accuracy of endoscopic retrograde cholangioscopy (ERC). ERC alone was less accurate in the diagnosis of ITS compared with POCS with or without biopsies, at 79.5 % versus 100 %, respectively. The extent of ITS could not be accurately assessed with ERC (accuracy 21.4 %), whereas POCS showed a diagnostic accuracy of 76.9 %, and adding mapping biopsy to POCS increased the diagnostic accuracy to 100 %. One could argue that these results are somewhat biased because 6 out of 44 patients (13.6 %) were excluded from analysis because the cholangioscope could not be passed through the strictured tumor site. However, this should not be regarded as a major issue. Stricture dilation of the tumor site with a 4-mm balloon, which was not done in this series, should allow passage of the 3.1-mm cholangioscope in the majority of patients. The composition of patients included in the study is worthy of remark. One would assume that patients with either proximal extrahepatic bile duct cancer or true hilar cholangiocarcinoma would benefit most from mapping the extent of malignant infiltration. In this series of 44 patients however, only three patients had their main tumor located in the right and/or left hepatic bile duct and for ten it was in the superior (proximal) extrahepatic bile duct. The latter patients had mid or low common bile duct tumors.

To my mind there is no doubt that the future belongs to cholangioscopy with regard to diagnosing or refuting biliary malignancy and to assess the extent of disease. Brushings and cholangiographically guided biopsies, whatever additional techniques are applied, simply can never meet the critical precondition for optimal cytological or histological diagnosis, that is, optimal tissue sampling. Intraluminal endoscopy allows visual inspection of suspicious tissue and adjacent areas. There is no apparent reason to assume that the optimal strategy for diagnosing and mapping intrabiliary cancer differs from that for early neoplasia in Barrett’s esophagus or for early gastric cancer, with the possible addition of techniques, including chromoendoscopy and narrow band imaging, that guide targeted biopsy by enhancing the visual distinction between normal and abnormal tissue [5] [6] [7].

A more widespread use of this technology beyond a few centers of excellence is dependent on several factors. Regular use of cholangioscopic instruments would be endorsed when this can be done at a reasonable cost per procedure. Such regular use would promote operator experience and hence the yield. Importantly, operator experience, especially in the case of mapping the extent of hilar tumors intrahepatically, entails more than using such devices a few times per year: it requires knowledge of hepaticobiliary anatomy and its variations and a close working relationship with hepatopancreaticobiliary surgeons. From a technical point of view, the cholangioscope must have excellent video optics to allow for optimal inspection of the mucosal surface, intraluminal maneuverability of the scope tip in two dimensions to facilitate targeted biopsy, and the presence of a working channel to pass a biopsy forceps. Moreover, it should be slim enough to allow passage through the stricture site albeit sufficiently robust for the instrument not to require return to the manufacturer for costly repair after every few investigations.

Various manufacturers offer devices and accessories for accomplishing direct cholangioscopy and several techniques have been described. Some investigators use regular ultraslim gastroduodenoscopes, for introduction into the biliary tree (after adequate papillotomy) either over a guide wire [8] or by using an intraductal balloon for fixation [9]. This is a valuable and cost-saving solution for treating biliary stones and acquiring biopsies for the diagnosis of cancer. It is less suited for mapping the extent of disease proximal from the stricture because of the scope size. The ”mother–baby“ scope concept involves introduction of a dedicated 3.1-mm cholangioscope through the working channel of a side-viewing endoscope. The optics are excellent and the baby scope is sufficiently narrow to be passed through strictures (with or without prior balloon dilation) [10]. A semidisposable solution is the Spyglass system which includes a reusable 6000-pixel fiber optic probe that is introduced through a disposable 10-Fr delivery catheter which also includes an accessory channel and channels for irrigation [11].

The cholangioscopy future looks bright! Instruments for intraluminal biliary endoscopy have evolved from ”toys for boys“ to devices that can potentially have an impact on the differential diagnosis of indeterminate strictures and the treatment of resilient bile duct stones. It is only a matter of time before cholangioscopy becomes a widespread and fully fledged diagnostic and therapeutic tool in the care of patients with pancreaticobiliary disease.

Competing interests: None

References

• 1 Jahng A W, Chung D, Pham B. et al . Staining for intracytoplasmic lumina and CAM5.2 increases the detection rate of bile duct cancers.  Endoscopy. 2009;  11 965-970
• 2 Kawakami H, Kuwatani M, Etoh K. et al . Endoscopic retrograde cholangiography versus peroral cholangioscopy to evaluate intraepithelial tumor spread in biliary cancer.  Endoscopy. 2009;  11 959-964
• 3 Nakanishi Y, Zen Y, Kawakami H. et al . Extrahepatic bile duct carcinoma with extensive intraepithelial spread: a clinicopathological study of 21 cases.  Mod Pathol. 2008;  21 807-816
• 4 Igami T, Nagino M, Oda K. et al . Clinicopathologic study of cholangiocarcinoma with superficial spread.  Ann Surg. 2009;  249 296-302
• 5 Hoffman A, Kiesslich R, Bittinger F. et al . Methylene blue-aided cholangioscopy in patients with biliary strictures: feasibility and outcome analysis.  Endoscopy. 2008;  40 563-571
• 6 Itoi T, Sofuni A, Itokawa F. et al . Peroral cholangioscopic diagnosis of biliary-tract diseases by using narrow-band imaging (with videos).  Gastrointest Endosc. 2007;  66 730-736
• 7 Brauer B C, Fukami N, Chen Y K. Direct cholangioscopy with narrow-band imaging, chromoendoscopy, and argon plasma coagulation of intraductal papillary mucinous neoplasm of the bile duct (with videos).  Gastrointest Endosc. 2008;  67 574-576
• 8 Larghi A, Lecca P G, Ardito F. et al . Evaluation of hilar biliary strictures by using a newly developed forward-viewing therapeutic echoendoscope: preliminary results of an ongoing experience.  Gastrointest Endosc. 2009;  69 356-360
• 9 Moon J H, Ko B M, Choi H J. et al . Intraductal balloon-guided direct peroral cholangioscopy with an ultraslim upper endoscope (with videos).  Gastrointest Endosc. 2009;  70 297-302
• 10 Shah R J, Langer D A, Antillon M R, Chen Y K. Cholangioscopy and cholangioscopic forceps biopsy in patients with indeterminate pancreaticobiliary pathology.  Clin Gastroenterol Hepatol. 2006;  4 219-225
• 11 Chen Y K. Preclinical characterization of the Spyglass peroral cholangiopancreatoscopy system for direct access, visualization, and biopsy.  Gastrointest Endosc. 2007;  65 303-311